Inhibition of Sphingosine Kinase 1 Ameliorates Angiotensin II-Induced Hypertension and Inhibits Transmembrane Calcium Entry via Store-Operated Calcium Channel

Abstract

Angiotensin II (AngII) plays a critical role in the regulation of vascular tone and blood pressure mainly via regulation of Ca²⁺ mobilization. Several reports have implicated sphingosine kinase 1 (SK1)/sphingosine 1-phosphate (S1P) in the mobilization of intracellular Ca²⁺ through a yet-undefined mechanism. Here we demonstrate that AngII-induces biphasic calcium entry in vascular smooth muscle cells, consisting of an immediate peak due to inositol tris-phosphate-dependent release of intracellular calcium, followed by a sustained transmembrane Ca²⁺ influx through store-operated calcium channels (SOCs). Inhibition of SK1 attenuates the second phase of transmembrane Ca²⁺ influx, suggesting a role for SK1 in AngII-dependent activation of SOC. Intracellular S1P triggers SOC-dependent Ca²⁺ influx independent of S1P receptors, whereas external application of S1P stimulated S1P receptor-dependent Ca²⁺ influx that is insensitive to inhibitors of SOCs, suggesting that the SK1/S1P axis regulates store-operated calcium entry via intracellular rather than extracellular actions. Genetic deletion of SK1 significantly inhibits both the acute hypertensive response to AngII in anaesthetized SK1 knockout mice and the sustained hypertensive response to continuous infusion of AngII in conscious animals. Collectively these data implicate SK1 as the missing link that connects the angiotensin AT_1A receptor to transmembrane Ca²⁺ influx and identify SOCs as a potential intracellular target for SK1.

Sphingosine 1-phosphate (S1P) has emerged as a potent bioactive lipid with established roles in myriad cellular responses including cell proliferation/DNA synthesis, survival, migration, angiogenesis, and intracellular Ca²⁺ mobilization (1, 2). S1P is produced by phosphorylation of sphingosine by two sphingosine kinase (SK) isoforms: SK1 and SK2. SK1 and SK2 have been reported to fulfill distinct functions. Although SK2 mainly plays a role in apoptosis and cell migration, SK1 stimulates proliferation and survival, and only SK1 has been implicated in Ca²⁺ mobilization. S1P appears to have both extracellular and intracellular sites of action as it acts in an autocrine/paracrine manner to stimulate G protein-coupled S1P_(1–5) receptors present on the cell surface and functions as an intracellular second messenger to stimulate mitogenesis (3) and Ca²⁺ mobilization (4). Whereas the autocrine/paracrine signaling pathway of S1P is well defined, the molecular mechanisms behind the intracellular action of S1P are poorly understood due to scarcity of information regarding intracellular target molecules.

The molecular mechanisms that mediate the cardiovascular response to angiotensin II (AngII) stimulation have been the subject of intense research efforts. AngII mediates a host physiological and pathological responses including vasoconstriction and water retention, increases in renal tubular sodium reabsorption, decreases in endothelial function, and stimulation of connective tissue deposition (5). These effects are mediated via complex, interacting signaling pathways involving stimulation of phosphorlipase (PL)-C and Ca²⁺ mobilization, activation of PLD, PLA2, protein kinase C, MAPKs, and stimulation of gene transcription. AngII 1A (AT_1A) receptor (AT_1AR) activation stimulates PLC-dependent hydrolysis of phosphatidylinositol 4,5-bisphosphate to generate two second messengers: inositol 1,4,5-tris-triphosphate (IP₃) and diacylglycerol. IP₃ binds to receptors in the endoplasmic reticulum (ER) and triggers the release of Ca²⁺ leading to a rapid increase in intracellular Ca²⁺. The initial IP₃-dependent Ca²⁺ peak is followed by a slower transmembrane Ca²⁺ entry that leads to a long-lasting plateau phase (5, 6). Despite the increasing knowledge of the role of various signaling proteins in AngII-induced cardiovascular diseases, the role of SK1/S1P pathway in AngII-mediated vascular dysfunction such as hypertension is not well defined.

The goal of this study was to define the role of the SK1/S1P axis in AngII-dependent Ca²⁺ mobilization and the contribution of SK1 to AngII-dependent hypertension. Using complementary pharmacological, molecular, and genetic approaches, we demonstrate that the SK1/S1P axis mediates AngII-dependent transmembrane Ca²⁺ influx via regulation of store-operated calcium entry (SOCE). Functionally we find that genetic deletion of SK1 moderates both acute and chronic AngII-induced hypertension.

Materials and Methods

Materials

Tissue culture medium, fetal bovine serum, and penicillin/streptomycin were from Invitrogen. FuGENE 6 was from Roche Diagnostics. Double-stranded small interfering RNAs (siRNAs) were purchased from Xeragon. GeneSilencer transfection reagent was from Gene Therapy Systems. Primers for real-time PCR were from Integrated DNA Technologies. RNeasy kits were from QIAGEN Corp, and iScript cDNA synthesis kits and iQ SYBR Green Supermix kits were from Bio-Rad Laboratories). SK&F96365 (SK&F), nifedpine (Nifed), bis-(o-aminophenoxy)-ethane-N,N,N′,N′-tetraacetic acid, tetra(acetoxymethyl)-ester, U73122, thapsigargin (Tg), EGTA, and ionophore A23187 were from Sigma-Aldrich. VPC23019 (VPC), JTE013, and caged S1P (cS1P) were from Avanti Polar Lipids. D-erythro-S1P and dimethylsphingosine (DMS) were provided by the Medical University of South Carolina (MUSC) Lipidomics Core (Department of Biochemistry, MUSC, Charleston, South Carolina), and SK1.I (BML-258) [N-methyl-5-(4-pentylphenyl)-2-aminopent-4-en-1,3-diol] was from Enzo Life Science, Inc. Rabbit polyclonal antihuman SK1 and IgG were from Exalpha Biologicals, and phospho-SK1 (Ser225) was from ECM Biosciences.

Animals

All of the experimental protocols were approved by the Institutional Animal Care and Use Committee at MUSC and Georgia Health Science University. Sprague Dawley rats and mice were housed and handled in the division of laboratory animal resources facility under MUSC and Georgia Health Science University guidelines. Mice were maintained under controlled conditions of humidity (50% ± 10%), light (12 h light, 12 h dark cycle) and temperature (23°C ± 2°C). SphK1 homozygous knockout mice of the 129_SV-C57BL/6 background, a kind gift from Dr Richard L. Proia (National Institute of Diabetes and Digestive and Kidney Diseases/National Institutes of Health, Bethesda, Maryland), were backcrossed to C57BL/6 wild-type (WT) mice at least 10 times. WT mice and Sprague Dawley rats were purchased from Charles River Laboratories.

Blood pressure (BP) measurement in AngII-induced acute hypertension

Two groups of 8-week-old SphK1^−/− (n = 5) and age-matched C57BL/6 male (WT) (n = 6) mice were prepared for the acute study under isoflurane anesthesia (5% induction and 1%–2% maintenance). Once anesthetized, the mouse was placed on a thermostatically controlled, heated table to maintain body temperature at 37°C. Two polyethylene catheters were inserted into the right jugular vein for the infusion of 1% BSA in 0.9% NaCl through the catheters to maintain fluid volume and for the infusion of agents (total infusion rate of 0.2 mL/h). The left femoral artery was catheterized to continuously monitor BP through a pressure transducer coupled to a BP recorder (DigiMed). The urinary bladder was cannulated through an abdominal incision to allow collection of urine. At the completion of the surgery, animals received a bolus infusion of 0.2 mL of 1% BSA in 0.9% NaCl followed by a 45-minute stabilization period. At the end of the stabilization period, 15 minutes of base line BP was recorded, then AngII (1 μg/kg · min in 1% BSA in 0.9% NaCl, 0.1 mL/h) was infused into the jugular vein. After a 3-minute equilibration period, BP was recorded each minute for a further 57 minutes. After the cessation of AngII infusion, BP was recorded for a further 15 minutes.

Induction of chronic AngII-dependent hypertension in conscious mice

In two groups of 10- to 12-week-old male SphK1^−/− (n = 6) and age-matched C57BL/6 mice (n = 6), a Data Sciences International telemetry catheter was placed in the left carotid artery under isoflurane anesthesia and using aseptic technique. The transmitter body was routed to the scapular region. After recovery, mice were transferred to the Laboratory Animal Services facility in a room with a 12-hour light cycle. Mice were housed individually in standard mouse shoebox cages. Each cage was placed on a Data Sciences International receiver that was matched via the software to that mouse's implanted telemetry unit. After 3 days of control measurements, hypertension was induced by AngII infusion (600 ng/kg · min) via an osmotic minipump (Alzet; model 1002) for 2 weeks. BP signals were collected at 500 Hz, for 4 seconds each minute, 19 hours per day. Daily mean arterial pressure (MAP) data were separated into day and night averages to show circadian rhythm.

Cell isolation and culture

Primary rat and murine aortic vascular smooth muscle cells (VSMCs) were isolated as previously described (7). Human embryonic kidney 293 (HEK293) cells were obtained from the American Type Culture Collection. Freshly isolated VSMCs and HEK293 cells were maintained in minimum essential medium (Lonza) supplemented with 10% fetal bovine serum and 1% antibiotic/antimycotic solution (Sigma Chemical Co). Cells were fed every 2 days and subcultured upon reaching 90% confluence. Prior to each experiment, cells were seeded into multiwell plates as appropriate and incubated for 24–48 hours in serum-free growth medium supplemented by 0.1% BSA and 1% antibiotic/antimycotic solution. All experiments using rat VSMCs were performed between passages four and nine, whereas that of murine VSMCs were performed on the second passages.

siRNA down-regulation

Proteins expressions were down-regulated using sequence-specific siRNAs targeting human (h) SK1 (5′-GGCCCAGCUGCCUAUGUAATT-3′ and 5′-UUACAUAGGCAGCUGGGCCCA-3′); hS1P₁ receptor (5′-GGAGUAGCCUACACAGCUATT-3′ and 5′-UAGCUGUGUAGGCUACUCCTG-3′); hS1P₂ receptor (5′-CCCAGACCUAGGCUAAUAAACGGTT-3′ and 5′-UUGGGUCUGGAUCCGAUUAUUUGCCAA-3′); hS1P₃ receptor (5′-GGUGGCCAACCACAACAACTT-3′ and 5′-GUUGUUGUGGUUGGCCACCTT-3′); scrambled siRNA sequences (5′-ACGUGACACGUUCGGAGAAAdTdT-3′ and 5′-UUCUCCGAACGUGUCA CGUdTdT-3′) and were used as negative controls. HEK293 cells were seeded in collagen coated 10-cm dishes at a density of 2 × 10⁵ cells per dish 24 hours before transfection. Cells were transfected using GeneSilencer siRNA transfection reagent (Gene Therapy Systems) according to the manufacturer's protocol. The efficiency of the knockdown was determined by quantitative real-time PCR for hSK1, hS1P_1–3 receptor mRNA and immunoblotting for hSK1 protein 48 hours after transfection.

Immunoblotting

HEK293 cells were plated onto poly-D-lysine-coated, 12-well multiwell plates and serum deprived overnight. Cells were lysed in Laemmli sample buffer. Samples containing 20 μg of protein/lane were resolved by SDS-PAGE and transferred to polyvinylidine difloride membranes. Blots were blocked in 4% BSA, probed with anti-SK1 or antiphospho-SK1 IgG, and then horseradish peroxidase-conjugated polyclonal donkey antirabbit IgG. Immune complexes were visualized by enzyme-linked chemiluminescence and quantified using a Fluor-S MultiImager. Blots were exposed to stripping buffer and then reprobed with antiglyceraldehyde-3-phosphate dehydrogenase (GAPDH) IgG followed by horseradish peroxidase-conjugated IgG.

Quantitative RT-PCR

The total cellular RNA was isolated using the RNeasy kit (QIAGEN Corp) according to the manufacturer's instructions. cDNA was prepared from 1 μg of total RNA with A₂₆₀/A₂₈₀ greeter than 1.8 using the iScript cDNA synthesis kit (Bio-Rad Laboratories) per the manufacturer's instructions. Quantitative real-time PCR was performed with an iCycler 1Q real-time detection system using the iQ SYBR Green Supermix kit (Bio-Rad Laboratories). Oligonecleotide primer sequences were as follows: hSK1, 5′-CTGATGCACGAGGTGGTGAAC-3′ and 5′-GTGACCTGCTCATAGCCAGCA-3′; hAT1a receptor, 5′-GGTATCGATCAATGCTGA-3′ and 5′-GCTGTCTACACAGCTATGG-3′; rat (r) S1P₁ receptor, 5′-CTCTCCGCAGCCAAGTCTCC-3′ and 5′-GTCCTTCTCCACTCCGATGTTCA-3′; rS1P₂ receptor, 5′-ACACTACAATTACACCAAGGA-3′ and 5′-GCGATTAGCACCAGAAGG-3′; rS1P3 receptor, 5′-GCTGTGAAGATGCTGATGAG-3′ and 5′-GGATGTGCTGGCTAATTGC-3′; hS1P₁ receptor, 5′-TTCATTCTCATCTGCTGCTTTATC-3′ and 5′-GGGTGGGTGGAATTTCTTGG-3′; hS1P₂ receptor, 5′-CATCAAGAGTCCCAAAGTCCT-3′ and 5′-CCATTCCTCATTCTCAGACCTC-3′; hS1P3 receptor, 5′-GCCGTGAAGATGCTCATGC-3′ and 5′-ACAATCTCCCTGACTGCTCTAC-3′; hGAPDH, 5′-CTGAGTACGTCGTGGAGTC-3′ and 5′-AATGAGCCCCAGCCTTC-3′, and rGAPDH, 5′-AAGTTCAACGGCACAGTCAAG-3′ and 5′-CATACTCAGCACCAGCATCAC-3′. Real-time PCR results were analyzed using Softmax Pro software (Molecular Devices Corp). Protein expression data were normalized to expression of GAPDH as an endogenous control.

Assays of SK activity and C17-sphingosine labeling

These were performed as previously described (8).

Determination of levels of S1P and C17-S1P by liquid chromatography/mass spectrometry analysis

S1P and C17 S1P were analyzed using a Thermo Finnigan TSQ7000 triple quadrupole mass spectrometer, operating in a multiple reaction monitoring positive ionization mode as previously described (9).

FLIPRTETRA assay of intracellular Ca²⁺ level

HEK293-AT_1AR cells and quiescent VSMCs were seeded on black well/clear bottom 96-well plates at a density of 20 000 cells/well and serum starved in serum free MEM (0.1% BSA, 1% penicillin/streptomycin) overnight or for 48 hours, respectively. After starvation, cells were incubated with calcium-sensitive fluorescent dye (FLIPR calcium 5 assay kit; Molecular Devices) for 1 hour at 37°C in the presence or absence of inhibitors and then washed twice with PBS. Stimulation is carried out on the fluorometric imaging plate reader, FLIPRTETRA, an instrument with 488 nm excitation wavelength as we have previously described (10). Acquired data were analyzed using system software.

Photolysis assay by multiphoton confocal microscopy

This was performed as described (11). Rat VSMCs grown in 35-mm dishes were incubated with calcium 4 dye for 45 minutes and then loaded with caged substrate (intracellularly) for 15 minutes or directly into the buffer (outside loading) as a control. Flash photolysis of caged substrate was conducted by brief light pulses of the 360 nm light at 20% intensity for 4 seconds and focused uniformly throughout the field of view. Fluorescence images were collected every 5 seconds, and images presented are from one microscopic field. All experiments were performed three times in at least 10 microscopic fields in each dish with similar results.

Statistical analysis

Cell culture experiments were each performed at least triplicate, and error bars represent SD. Statistical analyses were performed by a Student's t test. P values are defined in the figure legends. For in vivo studies in WT-treated and SK1^−/− mice, data are expressed as means ± SEM. In the acute studies, one-factor ANOVA with repeated measures was used to test the difference in baseline MAP between the two groups. The effect of AngII on BP was tested using an unpaired t test. To determine the overall effect of AngII on BP for the two groups, the areas under the curve (AUC) for the 55 minutes of BP data obtained during the AngII infusion was determined by subtracting the mean value of the two periods prior and after the AngII infusion from the total area during the AngII infusion. The calculated areas were compared using the Mann-Whitney U test. For the chronic in vivo study, data were analyzed using two-factor ANOVA with repeated measures followed by a Bonferroni post hoc test for multiple comparisons.

Results

AngII stimulation induced activation of SK1 and production of S1P

We first tested whether AngII treatment induces SK activation by measuring SK activity in rat VSMC lysates using the radiolabeled assay for SK activity. As shown in Figure 1A, stimulation of rat VSMCs with either AngII or phorbol myristate acetate for 5 minutes produced a significant increase in SK activity (45.12 ± 0.36 and 42.77 ± 1.19 pmol/min · mg protein, respectively, compared with a basal activity of 36.01 ± 2.25 pmol/min · mg protein). Time-course stimulation of AT_1AR-expressing HEK293 cells with AngII-induced phosphorylation of SK1 within 1 minute, an effect that persisted for up to 15 minutes (Figure 2B). As a complementary approach to confirm SK1 activation by AngII stimulation, we used in situ C17-sphingosine labeling. C17-sphingosine lacks one carbon in their nonpolar moiety compared with the natural C18-sphingosine and has been used recently as an indicator for SK1 activation (8, 12). As shown in Figure 2C, the stimulation of rat VSMCs with AngII for 5 minutes induced significant increase in C17-S1P production (picomoles per micromole phosphate) (0.878 ± 0.059, P > .03) compared with nonstimulated (NS) cells (0.699 ± 0.107). This effect was significantly inhibited (0.670 ± 0.029, P > .003) when cells were incubated with the SK1 specific inhibitor (BML-258) (SK1-I) for 30 minutes. To further confirm SK1 activation by AngII stimulation, we measured the S1P level in cell lysates in VSMCs isolated from WT and SphK1^−/− mice using tandem mass spectrometry. AngII stimulation for 5 minutes significantly increased S1P production in WT cells (picomoles per milligram protein) (0.268 ± 0.059, P > .02) compared with NS cells (0.171 ± 0.009), whereas the genetic deletion of SK1 significantly inhibited production of S1P in SphK1^−/− stimulated cells (0.166 ± 0.002, P > .005) compared with WT stimulated cells (Figure 1D).

Figure 1.

AngII stimulation induced activation of endogenous SK1 and S1P production. A, Serum-deprived rat VSMCs were treated with 100 nM AngII or 100 nM phorbol myristate acetate (PMA) for 5 minutes after which SK activity in whole-cell lysates was assayed as described. *, P < .05 vs NS. B, Serum-deprived HEK293 cells were stimulated with 100 nM AngII for indicated times, and activation of SK1 in whole-cell lysate samples was determined by immunoblotting with phosphorylation state-specific IgG. SK1 phosphorylation is expressed as fold increase above basal levels in unstimulated cells. A representative phospho-SK1 and basal GAPDH immunoblots are shown above a bar graph presenting mean ± SD of three independent experiments. *, P < .05 vs unstimulated. C, Serum-deprived rat VSMCs were incubated with 5 μM SK1-I for 30 minutes and 1 μM C17-sphingosine for 13 minutes and then stimulated with 100 nM AngII for 5 minutes after which C17–S1P in cell lysates were assayed as described. *, P < .05 vs NS, #, P < .05 vs stimulated. D, Serum-deprived murine VSMCs were stimulated with 100 nM AngII for 5 minutes and then lipids were extracted and assayed for S1P by quantitative mass spectrometry. Data shown represent the mean ± SD of three independent experiments. *, P < .05 vs NS, #, P < .05 vs stimulated WT.

Figure 2.

Inhibition of SK1 inhibits AngII-induced intracellular Ca2+ elevation. Serum-starved rat VSMCs incubated with Fluo-5 calcium sensitive dye and then exposed to 10 μM DMS and 5 μM SK1-I for 30 minutes and then stimulated with 100 nM AngII or 5 nM S1P (A and H, respectively) or 1 μM ionophore A23187 as a control (B). The change in intracellular calcium fluorescence was recorded in real time in the FLIPRTETRA at 470–495 excitation per 515–575 nm emission wave lengths. Data shown are normalized to basal and from a representative of three separate experiments. Serum-starved HEK293 cells were transfected with control scrambled siRNA (siRNA) or siRNA targeting the SK1 (siSK1) for 48 hours, and the levels of SK1 and GAPDH were determined by immunoblotting in whole-cell lysates (C). A representative of SK1 and GAPDH immunoblots are shown above bar graph depicting mean ± SD of three independent experiments. mRNA levels of SK1 (D), AT_1AR, and GAPDH (E) were determined by real time PCR. Serum-deprived cells transfected with SCR (F) or siSK1 (G) incubated with Fluo-5 calcium sensitive dye and then stimulated with 100 nM AngII and 1 μM ionophore. Data shown are normalized to its basal of three independent experiments produced same results. I, VSMCs from WT and SphK1^−/− mice incubated with Fluo-5 calcium sensitive dye and stimulated with 100 nM AngII. Data shown are one of three independent experiments (triplicate in each) gave the same results.

SK1 mediates AngII-induced intracellular Ca²⁺ elevation

To determine whether SK1 is involved in AngII-dependent intracellular Ca²⁺ mobilization, we measured real-time changes in intracellular Ca²⁺ levels using calcium-sensitive fluorescent dye. We first used the known pan SK inhibitor (DMS). Pretreatment of rat VSMCs with DMS for 30 minutes prior to stimulation with AngII-attenuated the initial phase of Ca²⁺ elevation and abolished the second phase (Figure 2A). A similar pattern was observed using the SK1 specific inhibitor SK1-I (Figure 2H), suggesting the involvement of SK1 in intracellular Ca²⁺ elevation by AngII. In contrast, exogenously applied S1P, the physiological agonist of G protein-dependent S1P receptors, provoked Ca²⁺ that was insensitive to SK1 inhibition compared with the control ionophore A23187 (Figure 2B).

We next used HEK293 cells stably expressing AT₁aR, which enabled us to efficiently down-regulate endogenous SK1 expression by siRNA. The down-regulation of hSK1 by specific siRNA reduced endogenous SK1 mRNA and protein abundance by 69% and 62%, respectively, and did not affect the AT₁aR expression levels, compared with scrambled control siRNA (SCR) (Figure 2, C–E). The down-regulation of SK1 markedly inhibited the initial phase of Ca²⁺ elevation and obliterated the second phase of AngII-induced intracellular Ca²⁺ elevation compared with cells treated with SCR (Figure 2, F and G). Calcium influx in response to the calcium ionophore, A23187, was unaffected. Notably, knockdown of SK2 mRNA and protein abundance by 76% and 61%, respectively, did not affect AngII-dependent Ca²⁺ elevation (data not shown), confirming the previously reported isoform-specific role of SK1 in AngII-dependent Ca²⁺ mobilization (4, 13). Furthermore, we used primary VSMCs freshly isolated from WT and SpkK1^−/− mice. Similar to the finding in Figure 2, A–H, using SK1 pharmacological inhibitors, genetic deletion of SK1 inhibited the initial phase and abolished the second phase of Ca²⁺ elevation, indicating that SK1 is required for sustained AngII-dependent elevation of intracellular Ca² (Figure 2I).

AngII-dependent activation of SOCE via the SK1-sensitive pathway

Upon stimulation, most excitable cells display a biphasic increase in cytosolic Ca²⁺ concentration. The initial transient increase reflects IP₃-mediated release of ER Ca²⁺, which is then followed by prolonged transmembrane extracellular Ca²⁺ influx. Although the immediate release of Ca²⁺ from intracellular stores is required for many of the acute responses to AngII, the transmembrane Ca²⁺ influx is attributed to regulation of various biological responses. To dissect the sources of the intracellular Ca²⁺ in our cell system, we first used U73122 to inhibit PLC and block IP₃-dependent Ca²⁺ release from intracellular stores. Pretreating rat VSMCs with U73122 for 30 minutes abolished Ca²⁺ elevation by AngII (Figure 3A), indicating that Ca²⁺ release from intracellular stores is required for the second phase of Ca²⁺ influx as previously reported (14, 15). To determine the channel responsible for extracellular Ca²⁺ influx, rat VSMCs were incubated with the voltage-dependent Ca²⁺ channel inhibitor, Nifed, or the store-operated Ca²⁺ channel (SOC) inhibitor SK&F for 30 minutes prior to stimulation with AngII. AngII-stimulated intracellular Ca²⁺ entry was not sensitive to Nifed, whereas SK&F significantly inhibited it, suggesting that transmembrane Ca²⁺ influx by AngII stimulation is predominantly through SOC (Figure 3B).

Figure 3.

AngII stimulation induced transmembrane Ca²⁺ entry via SOC and sensitive to SK1 inhibitor. Serum-starved rat VSMCs incubated with Fluo-5 calcium-sensitive dye were exposed to 5 μM U73122 (A) or 5 μM Nifed (B), and 5 μM SK&F for 30 minutes prior to stimulation with 100 nM AngII. Data shown are normalized to basal and from a representative experiment. Serum-starved rat VSMCs incubated with Fluo-5 calcium-sensitive dye were incubated with calcium-free buffer with 2 mM EGTA and then exposed to 5 μM DMS or (E), 5 μM Nifed, and 5 μM SK&F (D) for 30 minutes prior to stimulation with 3 μM Tg (C) or 100 nM AngII, and then 2 mmol/L calcium was added after 3 minutes. Serum-starved VSMCs from WT (F, G, and H) and SphK1^−/− (I and J) mice incubated with Fluo-5 calcium-sensitive dye were incubated with calcium-free buffer with 2 mM EGTA and then exposed to 5 μM SK1-I (G) or 5 μM Nifed, and 5 μM SK&F (H and J) for 30 minutes prior to stimulation with 3 μM Tg or 100 nM AngII (F and I), and then 2 mmol/L calcium was added after 3 minutes. The change in intracellular calcium fluorescence was recorded in real time in the FLIPRTETRA at 470–495 excitation per 515–575 nm emission wavelengths. Data shown are presented as δ-change in the intensity from a representative experiment.

SOCE is essential not only for replenishing intracellular stores but also for regulating various biological processes, including vascular tone (14, 16). To further assess the contribution of SOC to transmembrane Ca²⁺ entry upon AngII stimulation, we used a calcium depletion/calcium addition assay. Calcium released from ER has been reported to trigger extracellular Ca²⁺ influx via SOC (17). Thus, the elevation of intracellular Ca²⁺ in stimulated cells exposed to calcium free buffer will only represent Ca²⁺ released from intracellular stores. Any subsequent elevation of intracellular calcium occurring after addition of extracellular calcium will represent transmembrane Ca²⁺ entry through activated calcium channels (15). Incubation of VSMCs isolated from rat (Figure 3C) and WT and SpkK1^−/− mice (Figure 3, F–I) with Ca²⁺-free buffer (with 2 mM EGTA to deplete extracellular Ca²⁺) before stimulation with the specific inhibitor of sarco/ER ATPase, Tg, exhibited a rapid but transient rise in cytoplasmic Ca²⁺ due to the passive depletion of intracellular Ca²⁺ stores. Subsequent addition of 2 mM extracellular calcium resulted in elevation of intracellular Ca²⁺, confirming that this sustained phase is due to transmembrane Ca²⁺ influx. As shown in Figure 3, D and E, AngII stimulation of rat VSMCs produced a similar pattern of transient intracellular Ca²⁺ release followed by extracellular Ca²⁺ entry. Interestingly, this second phase of Ca²⁺ elevation was markedly inhibited by the SK1 inhibitor (DMS) or the SOC inhibitor (SK&F), whereas the voltage-dependent calcium channel inhibitor Nifed had no effect. Similarly, incubation of VSMCs from WT mice with SK1-I or SK&F prior to stimulation with AngII inhibited the second phase, whereas there was no effect using Nifed (Figure 3, G and H). Interestingly, the inhibitory effect of SK&F was not observed in SpkK1^−/− VSMCs, suggesting that SK1 is upstream of SOC (Figure 3J). These results confirm that the transmembrane Ca²⁺ influx after AngII stimulation is predominantly through SOCE and further suggest AngII-mediated SOC activation via SK1.

Extracellular S1P does not activate SOCE via plasma membrane (PM) S1P receptors

To assess whether G protein-coupled S1P receptors are involved in AngII-dependent Ca²⁺ mobilization, we established first the expression of S1P_1–3 receptors in rat VSMCs (Figure 4A). Rat VSMCs were incubated with SOC inhibitor, SK&F, or the S1P receptor antagonists VPC or JTE013 to block S1P_1–3 and S1P₂ receptors, respectively, for 30 minutes prior to stimulation (Figure 4B). AngII-induced intracellular Ca²⁺ elevation was significantly inhibited by SK&F and was not sensitive to S1P_1–3 receptor inhibitors, suggesting that intracellular Ca²⁺ elevation by AngII is independent of S1P receptors. To assess that, we used a siRNA approach to knock down S1P_1–3 receptors. Down-regulation of S1P receptors (68%, 71%, and 54% for S1P₁, S1P₂, and S1P₃ receptors; respectively) in AT_1AR-expressing HEK293 cells did not affect AngII-induced elevation of intracellular Ca²⁺ (Figure 4, C and D). We next applied S1P extracellularly to directly activate S1P receptors in VSMCs isolated from WT and SphK1^−/− mice. S1P stimulation-induced Ca²⁺ elevation was markedly inhibited by JTE013 (JTE) and VPC, indicating involvement of S1P_1–3 receptors in S1P-dependent Ca²⁺ mobilization (1). Interestingly, SK&F did not affect S1P-dependent intracellular Ca²⁺ elevation (Figure 4, E and F), which indicates that SOC is not downstream S1P receptors and further implicates SOC as a potential SK1/S1P intracellular target.

Figure 4.

AngII-dependent activation of SOCE is independent of S1P receptors. A, RNA was isolated from serum-starved rat VSMCs, and mRNA levels of S1P_1–3 receptors and GAPDH were determined by quantitative real-time PCR as described. B, Bar graph depicting the changes in calcium (AUC) estimated in rat VSMCs were seeded in 96-well plates with the same number in each well 48 hours prior to starvation. Cells then incubated with fluorescent dye were exposed to 5 μM VPC, 5 μM JTE, and 5 μM SK&F for 30 minutes prior to stimulation with 100 nM AngII. The change in intracellular calcium fluorescence was recorded in real time using Fluo-5 calcium-sensitive dye in the FLIPRTETRA at 470–495 excitation per 515–575 nm emission wave lengths. Data presented are mean ± SD of three independent experiments. *, P < .05 vs S1P stimulated. C, HEK293 cells were transfected with control SCR or siRNA targeting S1P₁R, S1P₂R, and S1P₃R. RNA was isolated 48 hours after transfection, and mRNA levels of S1P_1–3R and GAPDH were determined by quantitative real-time PCR as described. D, Bar graph depicting the changes in calcium (AUC) estimated in HEK293 cells transfected with scrambled siRNA or siRNA targeted to S1P₁R, S1P₂R, and S1P₃R were seeded in 96-well plates and stimulated with 100 nM AngII. Bar graph depicting the changes in calcium (AUC) estimated in Murine VSMCs isolated from WT (E) and SphK1^−/− (F) mice were seeded in 96-well plates with the same number in each well 48 hours prior to starvation. Cells then incubated with fluorescent dye were exposed to 5 μM VPC, 5 μM JTE, and 5 μM SK&F for 30 minutes prior to stimulation with 5 nM S1P. The change in intracellular calcium fluorescence was recorded in real time using Fluo-5 calcium-sensitive dye in the FLIPRTETRA at 470–495 excitation per 515–575 nm emission wave lengths. Data presented are mean ± SD of three independent experiments. *, P < .05 vs S1P stimulated.

Endogenous S1P induces transmembrane Ca²⁺ influx via SOCE

To test the alternative hypothesis that SOC is an SK1/S1P intracellular target, we used membrane-permeable UV light-sensitive cS1P and multiphoton confocal microscopy to photorelease S1P and simultaneously detect changes in intracellular Ca²⁺. We first established that a 360-nm wavelength laser flash for 4 seconds at 20% intensity was the optimal setting to release S1P from cS1P without affecting intracellular Ca²⁺ mobilization. We also established that incubating cells with SK&F for 30 minutes, or adding cS1P outside the cell without flashing (outside loading), did not affect Ca²⁺ mobilization in rat VMCS loaded with calcium sensitive dye. Outside loaded cS1P produced an immediate elevation of intracellular Ca²⁺ only when S1P was released by UV flashing to directly activate S1P receptors (Figure 5A). Consistent with the data shown in Figure 6, Ca²⁺ entry after the extracellular release of cS1P was insensitive to SOC inhibition. We next incubated rat VSMCs for with 1 μmol cS1P for 15 minutes (intracellularly) to load cS1P into cells as previously described (11), followed by washing to remove extracellular cS1P. Under these conditions, only intracellular S1P would be released upon UV flashing. As shown in Figure 5B, uncaging intracellular cS1P evoked immediate and sustained elevation of intracellular Ca²⁺ that was blocked by SK&F pretreatment (Figure 5B), indicating that transmembrane Ca²⁺ influx triggered by S1P is mediated by intracellular S1P via SOC and independent of S1P receptors.

Figure 5.

Intracellular S1P induces transmembrane Ca²⁺ influx via SOCE. A, Rat VSMCs grown in 35-mm dishes were starved for 48 hours and then incubated with calcium dye for 45 minutes. 1 μM caged S1P (cS1P) was added directly to media (out) (A) or incubated with 1 μM S1P cells (intracellularly, in) (B) for 15 minutes in the presence or absence of SK&F. Flash photolysis of caged substrate was conducted by brief light pulses of the 360-nm light at 20% intensity for 4 seconds and focused uniformly throughout the field of view. Fluorescence images were collected every 5 seconds, and images presented are from one microscopic field. Shown are representative confocal images from one of three independent experiments that gave similar results. Values in panel B are means ± SE for more than 10 microscopic fields/dish for three different experiments.

Figure 6.

A, Lack of SK1 attenuates acute AngII-induced elevation of BP. MAP in WT (n = 6) and SphK1^−/− mice (n = 5) infused with AngII (1 μg/kg · min) in anesthetized mice. Data are presented as means ± SEM. B, Genetic deletion of SK1 causes decrease in MAP in chronic AngII-induced hypertensive mice. MAP in WT and SphK1^−/− mice (n = 6 in each) infused sc with AngII 600 ng/kg · min by osmotic pump and monitored by telemetry in conscious mice. C, no AngII control; A, AngII infusion. Data are presented as means ± SEM. *, P < .05; **, P < .01 between groups.

Lack of SK1 attenuates acute AngII-induced elevation of BP

Our finding that inhibition or genetic deletion of SK1 inhibits AngII-dependent activation of SOCE suggests a role for SK1 in the regulation of vascular tone because the elevation of intracellular Ca²⁺ concentration is considered the principal process that initiates myogenic responses. Indeed, published reports indicated the requirement of SOC in arterial contraction and elevation of intracellular Ca²⁺ in VSMCs from hypertensive patients and from spontaneously hypertensive rats has been attributed to the increased responses to constrictor stimuli and augmented myogenic tone, the hallmark of arterial hypertension (17, 18). Thus, to test whether inhibition of SK1 would ameliorate AngII-induced hypertension, we first examined the BP response to acute infusion of AngII in WT and SphK1^−/− mice. As shown in Figure 6A, infusion of AngII-induced a similar initial increase over baseline in MAP in both groups of mice (13.3 ± 1.0 mm Hg and 13.9 ± 0.7 for the WT and SphK1^−/− mice at 6 minutes after the commencement of the infusion). The MAP then reset to a level that was sustained in the WT mice, whereas the prolonged response was attenuated in the SphK1^−/− mice (AUC for the AngII-induced increase in MAP: 3881 ± 89 and 3556 ± 124 mm Hg per 55 min for the WT and SphK1^−/− mice, respectively, P < .045). This finding indicates a role for SK1 in the homeostatic regulation of BP.

Genetic deletion of SK1 ameliorates AngII-induced chronic hypertension

To overcome the potential influence of anesthesia on BP in the acute study, we used telemetry to monitor changes in BP in WT and SphK1^−/− conscious mice infused sc with 600 ng/kg · min by a miniosmotic pump. As shown in Figure 6B, baseline MAP showed normal circadian rhythm in both groups and a decrease in the MAP of SphK1^−/− mice (111.3 mm Hg ± 3.27) compared with the WT control (119.4 mm Hg ± 2.14, P < .003), confirming our measurements in acutely perfused anesthetized mice. MAP rose within a day of implanting the AngII minipumps in the WT mice and continued to increase, whereas the MAP of SphK1^−/− mice remained near baseline for the full 14-day period (Figure 6B). This finding indicates an important role for SK1 in the initiation and/or progression of hypertension.

Discussion

Despite the accumulating evidence implicating the SK1/S1P pathway in regulation of intracellular Ca²⁺ mobilization, most attention to date has focused on the well-defined autocrine/paracrine signaling pathway via G protein-coupled S1P receptors. Although earlier studies have suggested an intracellular role for SK1/S1P in Ca²⁺ mobilization (4), the molecular mechanisms behind the action of intracellular SK1/ S1P have remained poorly understood due to scarcity of information regarding intracellular target molecules. Additionally, there has been little evidence that the regulation of Ca²⁺ mobilization via the SK1/S1P pathway plays a physiological role in the maintenance of vascular tone.

The present study addresses these concerns. Here we present evidence indicating the requirement of the SK1/S1P axis in AngII-dependent Ca²⁺ mobilization, implicating SOC as the elusive SK1/S1P intracellular target and suggesting a role for SK1 in AngII-induced hypertension. We first established that inhibition of SK1 inhibits AngII-induced intracellular Ca²⁺ elevation using pharmacological, molecular, and genetic approaches. Moreover, we demonstrated that AngII-stimulated transmembrane Ca²⁺ influx is mainly through SOC and that inhibition of either SOC or SK1 inhibited transmembrane Ca²⁺ influx with the same magnitude and independent of G protein-coupled S1P receptors. These findings suggest that SK1 is required for AngII-mediated SOCE and further suggest an intracellular role of SK1/S1P. The intracellular locus of S1P action is supported by data from primary VSMCs isolated from SphK1^−/− and WT mice, in which we demonstrated that, in contrast to AngII-mediated SOC-dependent Ca²⁺ influx, Ca²⁺ entry triggered by activation of S1P receptors was sensitive to S1P receptor antagonists but not SOC inhibition. In a direct test of whether intracellular release of S1P is sufficient to trigger SOCE, we found that SOC inhibition abolished elevation of intracellular Ca²⁺ triggered by intracellular photorelease of S1P from cS1P. This finding is in accordance with earlier reports indicating that S1P can trigger calcium responses either by an S1P receptor-mediated PLC/IP₃-dependent mechanism or by mediating intracellular Ca²⁺ mobilization independent of G protein-coupled receptors (reviewed in references 19 and 20). Subsequent studies provided more insight into the S1P effect on Ca²⁺ mobilization by using pertussis toxin, the G_i/o-protein inhibitor, to distinguish between extracellular effects via S1P receptors and intracellular actions. Although pertussis toxin inhibited the response of exogenous S1P, it did not prevent microinjection of S1P from releasing Ca²⁺ (21). Furthermore, endogenously generated S1P has been demonstrated to function as a positive modulator of SOCE, whereas exogenously applied S1P initiated calcium release from the ER in endothelial cells (22).

Two components of SOC that regulate transmembrane Ca²⁺ influx have been identified: STIM1, which is located in the ER and functions as an intracellular Ca²⁺ sensor, and Orai1, which is a component of Ca⁺² release-activated Ca⁺² channels (16, 23). Initiation of SOCE occurs when Ca²⁺ depletion provokes STIM1 oligomerization (24, 25), followed by translocation of these multimers to ER-PM junctions in which they activate Ca⁺² release-activated Ca⁺² channels (26). Our data clearly implicate intracellular S1P as a key player in SOC activation. Because two essential steps are required for SOC activation (STIM1 activation and translocation of STIM1 to the ER/PM junction to interact with Orai1), we hypothesize that SK1/S1P mediates activation SOCE either via activation of STIM1 or via regulation of STIM1 translocation and/or interaction with Orai1 at PM. Further investigation is needed to delineate the molecular mechanism underlying SK1-mediated SOCE.

Targeting calcium channels by pharmacological inhibitors was reported to cause modest inhibition in the IP₃-dependent phase of AngII-dependent elevation of intracellular Ca²⁺. This effect was attributed to the ability of these inhibitors to permeate PM and influence Ca²⁺ release from intracellular stores (27–29). Similar results were observed in the present study using calcium channel inhibitors as well as inhibition of SK1 by pharmacological and genetic approaches. This, however, may indicate that the subcellular localization of SK1 dictates distinct functional responses by acting on the sphingosine substrate that resides in different membrane compartments (30). Upon activation, SK1 translocates from the cytosol to intracellular membranes, in which it binds to sphingosine to generate S1P that contributes to both Ca²⁺ release from ER and transmembrane Ca²⁺ entry before acted upon by phosphatases and lyases (31). This is supported by our previous finding that SK1 activation by IGF2 was associated with translocation to PM and production of S1P (32) and an earlier study that demonstrated the effect of S1P on Ca²⁺ mobilization was selective to membrane preparations enriched with rough ER (33). However, the factors that dictate the subcellular targeting of SK1 to distinct cellular membrane and the molecular mechanism(s) through which SK1/S1P regulates SOCE require further investigation.

SOC contributes to the regulation of myogenic response, and disturbances in STIM1/Orai1 signaling have been implicated in vascular dysfunction, including hypertension. Increased expression of STIM1 and Orai1, associated with greater SOCE activation in VSMCs, has been reported in spontaneous hypertensive rats (17). Likewise, the SK1/S1P axis has been reported to regulate the myogenic responses by promoting both vasoconstriction responses mediated by RhoA/Rho-associated protein kinase pathways in VSMCs and vasorelaxation responses mediated by endothelial nitric oxide synthase/nitric oxide in endothelial cells (34–37). S1P induced vasoconstriction in canine basilar arteries (38), in rodent cerebral arteries (39, 40), and in rat mesenteric and renal arteries (41, 42). Overexpression of SK1 caused increase in resting tone and myogenic responses isolated resistance arteries, whereas overexpression of the dominant-negative SK1^G82D abolished development of tone and myogenic responses (35). Furthermore, the SK inhibitor (DMS) decreased vasomotor responses in isolated mice basilar arteries (36), suggesting SK1 as a mediator to vasomotor responses. In contrast to these findings, S1P has been shown to induce endothelial nitric oxide synthase-dependent vasorelaxation in epinephrine-preconstricted rat or mice mesenteric arteries and has little or no effect on aorta, carotid, and femoral arteries (43). SK mediated AngII-dependent production of endothelial nitric oxide that was blocked by DMS and thereby potentiated the AngII-induced contractile effect in isolated rat carotid arteries (44). This apparent discrepancy in the overall effect of SK1/S1P axis on vasoconstriction vs vasorelaxation using an ex vivo model of isolated blood vessels could be attributed to several factors. These factors include the specific vascular bed studied, S1P concentrations used, and the use of other vasoactive drugs in addition to other specific experimental conditions including the vascular disease model used and differences between animal species (45).

Recently the Kawamori group expanded previous studies (46, 47) to investigate whether inhibition of SK1 would induce cardiovascular risks using computerized noninvasive tail cuff system to measure BP in anesthetized mice injected ip with AngII (640 μg/kg body weight). Contrary to their hypothesis, this study concluded that the lack of SK1 not only did not aggravate but also ameliorated AngII-induced acute hypertension (48). These reports are in accordance with our finding that inhibition or genetic deletion of SK1 inhibited AngII-dependent activation of SOCE and significantly reduced the rise in BP by AngII infusion in in vivo animal models. It is worth noting that, although acute iv infusion of a low dose of AngII- induced a similar initial increase in MAP in both SphK1^−/− mice and WT controls (Figure 6A), in the continuing presence of the AngII, there was a small but significant attenuation of the BP response in the SphK1^−/− mice. This finding indicated a role for SK1 in the AngII-induced increase in BP. To test this, we examined the effect of genetic deletion of SK1 on chronic AngII-induced hypertension. The marked attenuation of the increase in MAP in the chronically AngII-treated SphK1^−/− mice provides support for the conclusion that SK1 plays a critical role in the development of AngII-induced hypertension. These findings implicate a key role for SK1 in the regulation of BP and may advance our understanding of the pathophysiology of hypertension.

In conclusion, our finding that SK1 mediates AngII-dependent transmembrane Ca²⁺ influx implicates SK1 as the missing link that connects AT_1AR to sustained transmembrane Ca²⁺ entry and identifies SOC as potential intracellular target for SK1. Furthermore, genetic deletion of SK1 significantly reduces BP in acute and chronic hypertension models. These findings may likely change the perception regarding the mechanism by which AngII regulates vascular tone and promotes hypertension and may identify a new therapeutic target for hypertension.